1. 程式人生 > >java字典樹(Trie)實現中文模糊匹配

java字典樹(Trie)實現中文模糊匹配

原理解釋:
java實現:https://blog.csdn.net/yuhk231/article/details/51539840
c實現:https://blog.csdn.net/qq_31175231/article/details/77827324
程式碼模板:缺點,只能檢索出在一個分支中的字首匹配內容

package com.xq.algorithm;
import java.io.BufferedReader;
import java.io.IOException;
import java.util.ArrayList;
import java.util.Arrays;
 
/*
 * 無資料結構設計下的蠻力中文鍵樹
 */
class TrieNode {
	public String value;
	public ArrayList<TrieNode> ptr = null;
	public TrieNode(String value) {
		this.value=value;
		ptr =new ArrayList<TrieNode>();
	}
}
 
public class TrieTree_1 {
	private static TrieNode root = null;
	ArrayList<String> searchResult=new ArrayList<String>();
	StringBuffer tempWord=new StringBuffer();
	int start=0;
	
	public TrieTree_1() {
		root = new TrieNode(null);
	}
	
	public void insert(String key) {
		TrieNode p = root;
		String tempWord;
		boolean contains;
		TrieNode tempNode;
		for (int i = 0; i < key.length(); i++) {
			tempWord=String.valueOf(key.charAt(i));
			contains=false;
			for(TrieNode tn:p.ptr){
				if(tn.value.equals(tempWord)){
					p=tn;
					contains=true;
					break;
				}
			}
			if(!contains){
				tempNode=new TrieNode(tempWord);
				p.ptr.add(tempNode);
				p=tempNode;
			}
		}
	}
	
	public ArrayList<String> search(String key) {  //模糊查詢就是這個方法,打個比方比如key是"ap",那麼ArrayList裡就有{"apple","application"}
		TrieNode p = root;
		String temp;
		boolean contains=false;
		for (int i = 0; i < key.length(); i++) {
			temp=String.valueOf(key.charAt(i));
			contains=false;
			for(TrieNode tn:p.ptr){
				if(tn.value.equals(temp)){
					p=tn;
					contains=true;
					break;
				}
			}
			if(contains){
				continue;
			}else{
				break;
			}
		}
		if(contains){
			if(!(p.ptr.isEmpty())){
				//查詢到關鍵字
				searchResult.clear();
				tempWord.delete(0, tempWord.length());
				tempWord.append(key);
				start=key.length();
				traverseTree(p);
			}else{
				//已經查詢到鍵樹的底部
				return null;
			}
		}else{
			//沒有查詢到相應關鍵字
			return null;
		}
		return searchResult;
	}
	
	private void traverseTree(TrieNode p){
		if(!(p.ptr.isEmpty())){
			for(TrieNode tn:p.ptr){
				tempWord.append(tn.value);
				start++;
				traverseTree(tn);
				start--;
				tempWord.delete(start,tempWord.length());
			}
		}else{
			searchResult.add(tempWord.toString());
		}
	}
	public static void main(String[] args) {
		TrieTree_1 chinese = new TrieTree_1();
		chinese.insert("中");
		chinese.insert("中國人");
		chinese.insert("中國");
		chinese.insert("中華人民");
		chinese.insert("中華人崛起");
		chinese.insert("中華上下五千年");
		ArrayList<String> list = chinese.search("中華");
		for (String string : list) {
			System.out.println(string);
		}
	}
}